Transiently Thermoresponsive Acetal Polymers for Safe and Effective

Subscriber access provided by University of Chicago Library

Article

Transiently Thermoresponsive Acetal Polymers for Safe and Effective Administration of Amphotericin B as Vaccine Adjuvant Simon Van Herck, Lien Van Hoecke, Benoit Louage, Lien Lybaert, Ruben De Coen, Sabah Kasmi, Aaron P Esser-Kahn, Sunil A. David, Lutz Nuhn, Bert Schepens, Xavier Saelens, and Bruno G. De Geest Bioconjugate Chem., Just Accepted Manuscript • DOI: 10.1021/acs.bioconjchem.7b00641 • Publication Date (Web): 26 Nov 2017 Downloaded from http://pubs.acs.org on November 28, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Bioconjugate Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Bioconjugate Chemistry

Transiently Thermoresponsive Acetal Polymers for Safe and Effective Administration of Amphotericin B as Vaccine Adjuvant Simon Van Herck,†,‡ Lien Van Hoecke,§,║ Benoit Louage,†,‡ Lien Lybaert,†,‡ Ruben De Coen,†,‡ Sabah Kasmi,†,‡ Aaron P. Esser-Kahn,┴ Sunil A. David,# Lutz Nuhn,†,‡ Bert Schepens,§,║ Xavier Saelens,§,║ Bruno G. De Geest *,†,‡ †Department ‡Cancer

of Pharmaceutics, Ghent University, Ottergemsesteenweg 460, 9000 Ghent, Belgium

Research Institute Ghent (CRIG), Ghent University, Belgium

§VIB-UGent

Center for Medical Biotechnology, Technologiepark 927, 9052 Ghent, Belgium

║Department

of Biomedical Molecular Biology, Ghent University, Ghent B-9052, Belgium.

┴Department

of Chemistry, University of California, Irvine, Irvine, California 92697, United States

#Department

of Medicinal Chemistry, University of Minnesota, Sixth Street SE, Minneapolis, Minnesota 55455, United States.

Corresponding author: [email protected]

ACS Paragon Plus Environment

1

Bioconjugate Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 27

ABSTRACT: The quest for new potent and safe adjuvants to skew and boost the immune response of vaccines against intracellular pathogens and cancer has led to the discovery of a series of small molecules that can activate Toll like receptors (TLRs). Whereas many small molecule TLR agonists cope with a problematic safety profile, Amphotericin B (AmpB) – an FDA-approved anti-fungal drug – has recently been discovered to possess TLR triggering activity. However, its poor aqueous solubility and cytotoxicity at elevated concentrations currently hampers its development as a vaccine adjuvant. We present a new class of transiently thermoresponsive polymers that in native state have a phase transition temperature below room temperature, but gradually transform into fully soluble polymers through acetal hydrolysis at endosomal pH values. RAFT polymerization afforded well defined block copolymers that self-assemble into micellar nanoparticles and efficiently encapsulate AmpB. Importantly, nanoencapsulation strongly reduced the cytotoxic effect of AmpB, but maintained its TLR triggering capacity. Studies in mice showed that AmpB loaded nanoparticles can adjuvant an RSV vaccine candidate with almost equal potency as a highly immunogenic oil-in-water benchmark adjuvant.


2

Page 3 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


INTRODUCTION The traditional vaccination paradigm relying on weakened or killed version of pathogens is failing for a wide number of viruses (e.g. HIV, RSV) and bacteria (e.g. mTB) as well as for therapeutic anti-cancer vaccination. 1

Therefore, novel strategies are endeavored based on subunit proteins and peptides to direct the immune

response against more conserved epitope domains or epitopes that are not involved in immune escape mechanisms.2 Furthermore, adjuvants have become essential components in many vaccines to modulate the interaction between antigen presenting cells, T cells and B cells, thereby skewing and amplifying the adaptive immune response.3 Several molecular adjuvants work by binding to pattern-recognition receptors (PRRs), like the Toll-like receptors (TLRs), leading to the secretion of immunostimulatory cytokines and typeI IFN.4,5 Over the years promising results have been achieved for small molecule agonists, in particular for TLR 4, 7, 8 and 9.6–12 Interestingly, a recent high content screening of 105 registered drug compounds by the David group has led to the identification of amphotericin B (AmpB) as activator of TLR 2 and 4.13 Owing to its well established safety record and FDA approved status, AmpB would be suitable for vaccine formulation. However, AmpB is poorly water-soluble and needs either covalent modification to render it more water soluble or encapsulation in polymeric and liposomal formulations.14–16 Despite its well established immunostimulatory properties17 to our knowledge no attempts have been made in formulating AmpB with the focus on its use as a vaccine adjuvant. With regard to the formulation of small molecule adjuvants there lies an important role for the delivery route to antigen presenting cells.18,19 Whereas water soluble molecular adjuvants hold the risk of systemic inflammation, nanoparticulate carriers can strongly limit this risk by directing the activity of molecular adjuvants to the immune-inducing sites in the draining lymph nodes 20 as recently shown by our group7 and others.21,22 Despite its FDA-approved state, an important constraint of AmpB is it cytotoxic action within a concentration window that is relevant for adjuvant activity. Therefore, we sought for a strategy to formulate AmpB as nanoparticles under benign conditions to alleviate cytotoxicity but retain its adjuvanticity. In this paper we develop a novel type of transiently thermoresponsive block copolymer23 nanoparticles to formulate AmpB (Scheme 1A) in the hydrophobic nanoparticle cavity (Scheme 1B). Such polymers reversibly form nanoparticles starting from a cooled polymer solution followed by a mild temperature increase not higher than ambient conditions, but lose this property – i.e. become fully water-soluble irrespective of temperature – as consequence of an alteration of the polymer side chain or backbone triggered by a physiologically relevant stimulus.23,24 The endosomal acidic pH of 5-5,5 is attractive to trigger such nanoparticle degradation and acetal/ketal chemistry has been used to serve this purpose. 25,26 Much research


3


Page 4 of 27

is done on polyacetal/ketal synthesis by step-growth polymerization.27–31 However, much less has been reported on acid-sensitive vinyl monomers that can be polymerized via controlled (radical) polymerization methods, with virtually no systems being reported which allow for degradation over an acceptably short timeframe within a physiologically relevant pH window.26,32,33

A B

C

Scheme 1. (A) Molecular structure of Amphotericin B. (B) Schematic representation of AmpB encapsulation in acidsensitive transiently thermoresponsive nanoparticles. (C) Synthesis and RAFT (block co) polymerization of HEAmTHP.

RESULTS AND DISCUSSION


4

Page 5 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


We engineered the water-soluble acrylamide 2-hydroxyethylacrylamide (HEAm), a readily available hydrophilic monomer, with an acid-sensitive hydrophobic acetal moiety using 3,4-dihydro-2H-pyran (DHP), known

from

protecting

group

chemistry,34

yielding

N-[2-[(tetrahydro-2H-pyran-2-yl)oxy]ethyl]-

acrylamide(HEAmTHP) (Scheme 1C). The applicability of the DHP group to form acid-degradable monomers through acetal linkage has been explored earlier by Klaikherd and co-workers35 on methacrylates (i.e. 2hydroxyethylmethacrylate (HEMA)), yielding acid-degradable, but fully hydrophobic, polymers. In our endeavors to design transiently responsive polymers, the use of acrylamides instead of (meth)acrylates, takes advantage of the presence of the amide moiety which gives the repeating units a more hydrophilic character, through hydrogen bonding with water molecules below the phase transition temperature (for the sake of simplicity denoted as Tcp, cloud point temperature, further in this paper) of the polymers. However, dehydration of the acrylamide side chains on its turn should provide a driving force for globule formation above a critical Tcp.24,36,37 Reactions were conducted at low temperature using equimolar amount of HEAm and DHP. Excess DHP resulted in lower yields due to formation of a side product, possibly formed by an “inverse demand” DielsAlder reaction between the alkene of DHP and the vinyl group of HEAm. NMR spectra of the purified monomer are provided in (Figure S1). MS analysis and the proposed structure of the aforementioned side product are shown in Figure S2. Degradability of the monomer in acidic buffer at pH 5 was confirmed by 1HNMR analysis (Figure 1A) showing the shift of the acetal proton (b) to the lactol proton (c). The pH-dependent hydrolysis rate is evaluated in aqueous buffers at pH 5 and 7.4 representing lysosomal and physiological pH conditions, respectively, by HPLC analysis (Figure 1B). Whereas a half-life of less than 24h was observed at pH 5, minimal hydrolysis was observed at pH 7.4, even after extended periods of time.


5


Page 6 of 27

Figure 1. Reaction scheme (A1) and 1H-NMR spectra (A2) of HEAmTHP before and after hydrolysis in deuterated buffer at pH 5. The most relevant H-peaks for hydrolysis are indicated in red. The peak annotated with “d” corresponds to the protons attached to the α-C of 5-hydroxypentanal which is only formed in very small amounts. (B) Degradation profiles in function of time of HEAmTHP and pHEAmTHP42 (P01) at pH 5 and pH 7.4 analyzed by HPLC or NMR. (C) Mean particle diameter (red) and scattering intensity (blue) profiles in function of temperature analyzed by DLS at 5 mg/mL in PBS for pHEAmTHP42 (P01). (D) Transmittance profiles of pHEAmTHP42 (P01) dispersions in function of temperature at different time points during hydrolysis at pH 5 (D1) and pH 7.4 (D2) analysed by turbidimetry. (E) Temperature induced self-assembly of P6 (p(HEA110-HEAmTHP47)) shown as average volume diameter (red) and scattering intensity (blue) in function of temperature measured by DLS in PBS. (F) pH dependent hydrolysis and disassembly of nanoparticles based on P6. Volume mean diameter (red) and scattering intensity (blue) as function of time at pH 5 (F1) and volume mean diameter (F2) at 2 different time points measured at pH 5 (red) and pH 7.4 (blue).

Homopolymers of HEAmTHP were synthesized by Reversible Addition-Fragmentation Chain Transfer (RAFT) polymerization using 2-(butylthiocarbonothioylthio)propionic acid (PABTC) as chain transfer agent (CTA), aiming at a degree of polymerization (DP) of 50 or 100 (Table 1). As an example, the 1H NMR spectra of the purified polymers are shown in Figure S4. Based on 1H NMR analysis of the monomer conversion a theoretical DP of 42, 66 was calculated for pHEAmTHP. For the sake of clarity, further on in this paper, the theoretical DP values will be mentioned in subscript when referring to a specific polymer. According to size


6

Page 7 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


exclusion chromatography (SEC) analysis, narrowly dispersed polymers were obtained, with dispersity (Ð) values of 1.12 and a number average molecular weight (Mn) of 8,6 and 13,4 kDa for pHEAmTHP42 (P01) and pHEAmTHP66 (P02), respectively. (Table 1, Figure S5). The thermoresponsive properties of the polymers were measured by dynamic light scattering (DLS) (Figure 1C and Figure S6) at a polymer concentration of 5 mg/mL in phosphate buffered saline (PBS; pH 7.4, 0.15 M NaCl) and revealed a cloud point temperature Tcp well below room temperature. The transiently thermoresponsive behavior properties – i.e. a gradual shift in Tcp upon hydrolysis of the pending THP moieties - of the polymers was investigated by turbidimetry on a colloidal suspension of pHEAmTHP (P01 or P02) at 37 °C (Figure 1D and Figure S7). A clear shift in Tcp was found which proves that pHEAmTHP loses its hydrophobic properties to become fully soluble irrespective of temperature.

Table 1. Polymer composition and properties of the synthesized (block co)polymers. No.

M:CTA

CTA [HEA]

conversion (%)

DPa

Mna (kDa)

Mnb (kDa)

Ð

[HEAmTHP]

P01

pHEAmTHP42

P02

pHEAmTHP66

PABTC

P03

pHEA78

PABTC

100:1

P04

pHEA110

PABTC

150:1

P1

p(HEA78-HEAmTHP19)

pHEA78

21:1

P2

p(HEA78-HEAmTHP38)

pHEA78

P3

p(HEA78-HEAmTHP53)

P4

p(HEA78-HEAmTHP70)

P5 P6

83.8

42

8.6

7.4

1.12

66

66

13.4

12.1

1.12

78

78

9.3

17.5

1.14

73

110

13.0

21.9

1.17

93

19

13.1

21.5

1.16

40:1

94

38

16.9

23.0

1.16

pHEA78

60:1

88

53

19.9

25.1

1.16

pHEA78

83:1

85

70

23.2

26.9

1.14

p(HEA110-HEAmTHP26)

pHEA110

30:1

87

26

18.6

27.3

1.17

p(HEA110-HEAmTHP47)

pHEA110

54:1

87

47

22.4

29.8

1.15

P7

p(HEA110-HEAmTHP70)

pHEA110

85:1

82

70

27.0

30.9

1.17

P8

p(HEA110-HEAmTHP87)

pHEA110

116:1

75

87

30.3

35.4

1.15

50:1 100:1

PABTC

Calculated based on 1H NMR spectroscopy data. bDetermined by SEC in DMAc using PMMA for calibration.

a

Amphiphilic block copolymers containing pHEAmTHP as responsive block were synthesized by RAFT polymerization as shown in Scheme 1. For the hydrophilic segment, 2-hydroxyethylacrylate (HEA) was selected, using a pHEA macroCTA approach, as it was – at least in our hands – not possible to obtain block copolymers using pHEAm as macroCTA. Neither was it possible to chain extend pHEAmTHP with HEAm. We attribute this to incompatibility issues between the respective monomers and the solvent system used for RAFT polymerization. Similar issues were also encountered recently within our laboratories and could also


7


Page 8 of 27

be resolved by combining acrylates with acrylamides.38 A pHEA macroCTA was synthesized (Figure S8 for 1H

NMR) with a targeted DP of 100 and 150, and the polymerization reaction was stopped before 80 %

conversion to ensure good chain end fidelity. 1H NMR analysis revealed a theoretical DP of 78 and 110, respectively. SEC analysis (Figure S10, red curves) confirmed the polymers to be narrowly dispersed, Ð of 1.14 and 1.17, with absence of tailing (Table 1). Next, the pHEA macroCTAs were used for the synthesis of a series of block copolymers (Table 1, Figure S9 and S10). The chain length of the hydrophobic block was varied to investigate its influence on the self-assembly behavior, temperature responsiveness, degradation rate and ability to encapsulate hydrophobic compounds. SEC analysis of these block copolymers revealed that all were obtained with a narrow dispersity, giving proof of well controlled block copolymerization (Table 1 and Figure S10). The influence of the hydrophobic chain length and total chain length on the self-assembly behavior was investigated in aqueous medium. An overview of these results is given in Table 2. Interestingly, all pHEAHEAmTHP block copolymers were able to self-assemble into small and defined micellar nanoparticles with low polydispersity and Z-average values ranging from 26 nm to 36 nm for polymers P1-P4 and from 28 nm to 42 nm for polymers P5-P8 at 37 °C in PBS. An increase in HEAmTHP chain length resulted in an increase in particle size (Figure S11). The influence of temperature on the self-assembly behavior of the block copolymers was measured by DLS (Figure 1E) where a strong influence of the polymer chain length on the phase transition temperature was observed, with shorter polymer chain lengths yielding a higher Tcp value. The critical association concentration (CAC) was determined via pyrene assay and also indicated a strong influence of the HEAmTHP content. Overall, a compelling correlation was seen between the hydrophobic chain length and the hydrodynamic diameter, the Tcp and CAC (Table 2, Figure S11), with decrease in CAC and Tcp upon increasing hydrophobic content. The latter, together with the ease of particle formation that proceeds via mere dissolution of the block copolymers in cold (i.e. below the Tcp) PBS followed by heating to room temperature, makes the HEA-HEAmTHP polymer combination an appealing system for drug formulation. The effect of the pH dependent acetal hydrolysis on the nanoparticle stability was investigated by DLS at pH values of 7.4 and 5. As shown in Figure 1F, the nanoparticles are stable at pH 7,4, but gradually disassemble in acidic medium (Figure 1F1), finally resulting in soluble unimers (Figure 1F2).


8

Page 9 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table 2. Overview of the supramolecular properties of synthesized pHEAx-HEAmTHPy block copolymers. H2O

PBS

25°C Z-average (nm)

1

5°C PdI

25°C

Tcp (°C)1

CAC (µg/mL)2

37°C

Z-average (nm)

PdI

Z-average (nm)

PdI

Z-average (nm)

PdI

P1

12.7 ±0.3

0.19

5.8 ±0.4

0.63

19.9 ±0.1

0.11

26.3 ±0.6

0.24

22

38,2

P2

14.1 ±0.3

0.21

7.4 ±0.7

0.53

27.5 ±0.3

0.05

29.2 ±0.5

0.03

10

23,0

P3

13.0 ±0.3

0.18

7.8 ±0.2

0.50

35.7 ±0.2

0.04

35.4 ±0.1

0.04

8

10,4

P4

29.4 ±4.9

0.40

7.8 ±0.3

0.31

41.3 ±1.1

0.06

36.1 ±1.6

0.06

5

8,47

P5

14.0 ±0.4

0.36

7.5 ±0.6

0.64

25.9 ±0.4

0.10

27.9 ±0.1

0.08

13

40,2

P6

14.2 ±0.2

0.20

7.9 ±0.3

0.72

32.0 ±0.1

0.07

32.5 ±0.1

0.02

9

16,9

P7

12.5 ±0.2

0.09

7.9 ±0.4

0.62

35.5 ±0.3

0.05

36.3 ±0.04

0.03

8

8,71

P8

18.3 ±0.3

0.18

47.5 ±11

0.33

54.8 ±0.8

0.13

42.4 ±0.3

0.01

5

5,53

Measured in PBS at 5 mg/mL; 2 Dilutions made in 0.1 M phosphate buffer, measurement done in triplicate

Next, we aimed at investigating the potential of our transiently responsive polymer system as a nanoscopic delivery system in a biomedical context. Initially, we assessed whether the block copolymer micelles could entrap a hydrophobic payload and deliver this to dendritic cells (DCs). DCs, as the most potent antigen presenting cells, are the prime target in the scope of formulating an adjuvant nanocarrier. 5,19 In first instance we made use of the hydrophobic fluorescent dye, 1,4-bis[1-cyano-2-(diphenylamino)phenyl)vinyl]benzene (DSB),39 as a model compound for physical encapsulation in order to visualize intracellular uptake. Encapsulation of DSB in pHEAx-HEAmTHPy nanoparticles was accomplished by solvent diffusion from acetone into PBS. Quantification of the amount of encapsulated DSB was done by UV-Vis spectroscopy after filtration to remove non-encapsulated precipitated dye and freeze drying (Figure S12). Control experiments were performed in only PBS without block copolymer while using P03 without polyHEAmTHP segment as a hydrophilic-only control. All pHEA-HEAmTHP block copolymers were able to encapsulate the hydrophobic DSB to a similar extent. Also, merely hydrophilic pHEA homopolymer could stabilize DSB, albeit to a much lesser extent. From these results, P6 was chosen to assess whether these block copolymers could increase the intracellular delivery of the hydrophobic DSB. For this purpose, DC2.4 (an immortalized mouse dendritic cell line) cells were pulsed with DSB formulated either in P6 nanoparticles or in P03 as a control. Subsequently, confocal microscopy was used to monitor the intracellular fate of the DSB. From Figure 2A, it is clear that DSB formulated with P6 leads to a punctuated intracellular fluorescence pattern (green dots), whereas no uptake of DSB could be observed in cells pulsed with DSB admixed with polyHEA. Note that both images were recorded and processed using identical settings.


9


Page 10 of 27

Encouraged by the observed superiority of P6 nanoparticle system in effectuating the intracellular uptake of a hydrophobic compound by DCs in comparison to free component, we use this block copolymer for the formulation of AmpB, which was the primary target compound in our study. In a first series of experiments, we screened different conditions to formulate AmpB using P6. Due to the limited solubility of AmpB in most organic solvents,14 solvent diffusion from DMSO into PBS was selected. We initially selected diluting a DMSO solution with PBS. Variations were made in the concentration of AmpB and polymer (P6), and in AmpB to polymer ratio (i.e. 1:10, 1:2 or 1:1). A graphical overview of these results can be found in Figure S15. Quantification of the encapsulated AmpB content was done by UV-Vis spectroscopy. For the latter, formulations were diluted with DMSO to release AmpB from the micelles, which yielded a typical multi-peak absorbance spectrum of AmpB shown in Figure S13. Encapsulation efficiency was found to increase with increasing polymer concentration. At lower polymer concentrations (1 mg/mL and 5 mg/mL) the 1:2 AmpB:polymer ratio yielded the best results regarding absolute encapsulated AmpB amount. This was however not the case for polymer concentrations at 10 mg/mL, as the 1:10 ratio proved to be most efficient. In this respect, the 1:10 ratio and a polymer concentration of 10 mg/mL was selected for further experiments based on the good encapsulation efficiency and absolute encapsulated concentration. Control experiments were performed in a similar fashion without block copolymer or only hydrophilic pHEA. In those cases, no AmpB was detected by UV-VIS after filtration indicating that amphiphilic block copolymers were necessary to efficiently solubilize AmpB. The above-elucidated optimal formulation conditions were used to evaluate the influence of hydrophobic chain length and total chain length of pHEA x-HEAmTHPy block copolymers on AmpB loading efficiency. Comparable results were obtained for encapsulation efficiency ranging between 35 % and 50 % (Figure S16), indicating that all polymers are suitable to prepare formulations. Overall, both P2 and P6 were found to be most suitable for formulating AmpB based on the polymer properties and encapsulation efficiency. The formulation method was further optimalized and scaled-up for P6. Finally up to 95 % of AmpB encapsulation was achieved by dropwise addition of a concentrated AmpB solution in DMSO into the polymer dispersion under vigorous stirring. The formulations were dialysed to remove the DMSO and freeze dried to have it in dry powderish form which could easily be redispersed in PBS to give a stable AmpBNP formulation (Figure S17).


10

Page 11 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2. (A) Confocal microscopy images of DCs pulsed at 37°C with DSB (green channel) formulated with P6 (A1) or P03 (A2). Cell membrane staining was done with AlexaFluor555 conjugated cholera toxin B (red channel) and cell nuclei staining was done with Hoechst (blue channel). Images are confocal xy section with the corresponding orthogonal xz and yz planes. Identical settings for excitation power, detection sensitivity and contrast were used for recording. Scale bar represents 10 µm. (B) In vitro cytotoxicity of AmpB-NP(P6), free AmpB and unloaded polymer (P6) on murine RAW 264.7 macrophages after 24h incubation. The graph shows the cell viability as measured by MTT assay. Cell viability of PBS treated cells was set as 100%. (n=6, mean + SD). (C) NF-κB induction in RAWBlue cells pulsed with a concentration range of AmpB, AmpB-NP (P6), and empty NP (P6) control. NF-κB activity presented relative to secreted embryonic alkaline phosphatase (SEAP) production detected by QUANTI-Blue assay. Note that these experiments were performed with nanoparticles (NP) composed of P6 (n=6, mean + SD). The black arrow highlights the drop in activity that is due to cytotoxicity.

To evaluate the adjuvant properties of AmpB loaded nanoparticles we firstly evaluated the in vitro cytotoxicity of our formulations in a murine RAW 264.7 macrophage cell line. The MTT assay indicated (Figure 2B) no cytotoxicity for both the polymer and the AmpB formulations over a concentration range from 0.1 – 30 µM. By contrast, AmpB in soluble form, i.e. diluted from DMSO in cell culture medium, was (highly) toxic with an LC50 value at approximately 4 µM. Note that this effect was not due to the presence of the minimal amount of DMSO used to primarily dissolve the AmpB, as a control experiment comprising only DMSO did not show


11


Page 12 of 27

any cytotoxicity. Moreover, the cytotoxic effect of AmpB as such has been reported earlier by others as well.40,41 Interestingly, AmpB formulated in P6 did not cause any cytotoxicity, at least not within the operated experimental window. The most commonly accepted mechanism of action and toxicity involves the formation of ion channels40,42,43 and is related to the physical state of the drug.14,44,45 It has been reported by various researchers that at low AmpB concentrations, where AmpB is in its monomeric form, it is less toxic towards mammalian cells. In contrast, above certain concentrations, when self-associated AmpB species form, it is nonselective and evokes cell lysis in both mammalian and fungal cells. 14,15,44 However, binding conclusions cannot be drawn from earlier reports studying this link due to presence of toxic solvents or surfactants in their experiments.14,45 The non-toxic behavior of the AmpB loaded nanoparticles could imply that the pHEAHEAmTHP block copolymers stabilize the AmpB in its monomeric form, thus making it non-toxic towards murine macrophages. Second, we tested the ability of the formulations to induce TLR activation using the RAW-Blue reporter cell line. These are murine derived RAW 264.7 macrophages that secrete an alkaline phosphatase (SEAP) upon activation of NF-κB, e.g. after TLR stimulation. A comparison is made between free AmpB and AmpB loaded micelles of block copolymer P6. Empty micelles were included as a control. NF-κB activity is presented as an increase in optical density, measured at 620 nm, relative to the negative control. The results in Figure 2C indicate that P6 based AmpB formulations can activate NF-κB. The absence of stimulation when only polymer is present shows that the activity is linked to AmpB released from the nanoparticle and not to the polymer. Free AmpB in its soluble monomeric form is at low concentration, below 1 µM, more potent in TLR activation than in micelle-encapsulated form. However, at concentrations above 1 µM, TLR activation of the soluble AmpB drops. Most likely this drop results from the detrimental effect of free AmpB on cell viability at concentrations above 1 uM (Figure 2B). The underlying reason for the reduced TLR agonistic activity of AmpB loaded NP in comparison with free AmpB is likely a more efficient triggering of the cell membrane bound TRL2/4 in soluble form than in micelle-encapsulated form.5,46 However, more importantly, micelleencapsulation allows for the use of higher AmpB concentrations, reaching overall a more potent TLR triggering (as the measured OD values of micelle-encapsulated kept on increasing above the value reached at 1 µM of soluble AmpB) while avoiding toxicity issues. Encouraged by these findings, the final aim of this study was to evaluate the potential of AmpB loaded nanoparticles as a vaccine adjuvant in vivo. To this end we prepared vaccine formulations by admixing nanoparticles based on P6 with an antigen derived from human respiratory syncytial virus (RSV). RSV is one of the leading causes of airway infections worldwide and is the leading cause of infant hospitalization.48


12

Page 13 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Although in most cases RSV causes mild upper respiratory tract infections, the virus can also spread the lower respiratory tract, which, in infants, can lead to bronchitis and increased risk of recurrent wheezing and allergic asthma at a later age.49 The inability of natural RSV infection to elicit protective immunity and the observation that individuals with RSV-neutralizing antibodies in their serum can be reinfected with RSV has hampered vaccine development. One RSV vaccine candidate, which is currently evaluated in a phase I trial, is based on the extracellular part of the small hydrophobic protein (SHe) of RSV.50 Interestingly, the mechanism of action of SHe-based vaccination does not rely on neutralizing antibodies but rather on binding of anti-SHe antibodies to the surface of infected cells and subsequent induction of antibody-mediated cell killing or phagocytosis. SHe is naturally poorly immunogenic and to explore its potential as a protective vaccine antigen, the immunogenicity of SHe has to be boosted. In our present study, we made use of a SHeKLH (KLH: keyhole limpet hemocyanin) fusion protein that was previously reported to elicit SHe-specific antibodies that can control RSV replication in a mouse model.50 DLS analysis showed no influence of the presence of the SHe-KLH protein on the colloidal stability of the nanoparticles (Figure S19). As well established adjuvant formulation, we opted for Sigma Adjuvant System (SAS), which is a commercially available oil-in-water emulsion based on squalene as oil phase stabilized with Tween80 as surfactant. Furthermore, it contains monophosphoryl lipid a (MPLA) from Salmonella minnesota and synthetic Trehalose dicorynomycolate (TDM). SAS is therefore comparable to the Ribi adjuvant and Freund’s adjuvant system, and often used for preclinical models, e.g. to assess the potency of new adjuvant systems against a very competitive benchmark. In our present experiments formulations were prepared by admixing KLH conjugated SHe-antigen with the respective adjuvants systems. She-KLH antigen only, antigen admixed with empty (i.e. non-AmpB loaded) nanoparticles and a PBS were included as controls. Note that it was impossible to prepare an antigen + free AmpB formulation due to the inability of AmpB to dissolve in aqueous medium at the required concentration for immunization. Cohorts of 5 BALB/c mice per group were immunized subcutaneously with one prime and one boost injection. Total anti-RSV SHe IgG antibody titers in serum were determined pre- and post-immunization and the dilution curves measured by a SHe-peptide based ELISA (enzyme linked immunosorbent assay) are depicted in Figure 3A-B. In addition, the end point titers are shown in Figure 3C. From these data, it is clear that antigen only and antigen formulated with empty nanoparticles are poorly immunogenic and only raise detectable antibody titers after the boost immunization. These findings also point out the low immunogenicity of the nanoparticles themselves. By contrast, antigen adjuvanted with SAS or AmpB-loaded nanoparticles induce robust antibody responses from the prime immunization onwards that are further increased upon the booster immunization. No significant differences


13


in SHe-specific antibody titers were observed between SAS and AmpB-loaded nanoparticles. This underscores the potency of the newly developed AmpB-loaded nanoparticles which does not involve bacteria cell wall components and are based on AmpB, an established and FDA approved drug molecule. To investigate whether the deployed immunization strategy could counteract RSV infection, mice were challenged with 106 plaque-forming units (PFU) of HRSV A2 followed by dissection of the lungs 5 days postinfection and measuring viral lung titers in an in vitro plaque formation assay.50 As depicted in Figure 3C, immunization with antigen adjuvanted with either SAS or AmpB-loaded nanoparticles significantly lowered RSV replication in the lung, again pointing out the excellent performance of our presently reported adjuvant system compared to a potent benchmark adjuvant.

108

A2

boost

Ag + AmpB-NP

1.5

Ag + SAS Ag +NP

1.0

Ag PBS

103

4

Ag +

0.0 101

104

PB S

0.5

105

102

103

104

105

Dilution

106

107

108

Ag g

+

N

SA

g

A

S

-N

107

p

106

P

105

Dilution

+

104

A

103

106

+

102

ns

A

0.0 101

****

5

S

PBS 0.5

**** ns

log10 lung RSV titer (PFU/lung)

Ag

B

PB

1.0

boost prime

Am pB -N P

Ag +NP

*** / **** *** / **** * / ns

Ag +S AS

Ag + SAS

107

Ag +

OD (450nm)

1.5

A3

NP

Ag + AmpB-NP

Ag

prime

total IgG endpoint titer (dilution)

A1

OD (450nm)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 27

A

m

P

B

g

Figure 3. Serum dilution curves after (A1) prime and (A2) boost immunization and (A3) end-point titers tested in a SHe peptide ELISA assay. (n=5) (B) Number of plaque forming units (PFU) in the lungs of immunized mice that had been challenged with RSV. (n=5) (*

Transiently Thermoresponsive Acetal Polymers for Safe and Effective

Recommend Documents